Vol. 397: 103–112, 2009 MARINE ECOLOGY PROGRESS SERIES Published December 17 doi: 10.3354/meps08322 Mar Ecol Prog Ser

Contribution to the Theme Section ‘Conservation and management of deep-sea corals and coral reefs’ OPENPEN ACCESSCCESS

Reproductive biology of the deep-sea pennatulacean Anthoptilum murrayi (Cnidaria, Octocorallia)

D. O. Pires*, C. B. Castro, J. C. Silva

Museu Nacional/Universidade Federal do Rio de Janeiro, Quinta da Boa Vista, s/n, São Cristóvão, Rio de Janeiro 20940-040, Brazil

ABSTRACT: Anthoptilum murrayi has been reported from the North Atlantic, northern Mid-Atlantic Ridge, Indian Ocean and in waters around New Zealand and Australia. Recently, this species was also recorded in deep waters off Brazil, southwestern Atlantic. It was from this region (13° to 22° S) that specimens were collected, in 1300 to 1799 m, to determine the reproductive biology of A. mur- rayi using histological methods. The colony polyparium was divided into 3 zones (distal, medial and basal) to evaluate differences in gamete development between zones; dissected polyps were exam- ined from the 3 zones to estimate fecundity. The species appears to display a continuous and long breeding activity rather than any seasonal reproductive pattern. Most oocytes were in the earliest stages of development and basal polyps presented the highest frequency of small eggs. The large mature oocytes (up to 1200 µm) indicate that A. murrayi produces lecithotrophic larvae. Females had 0 to 90 oocytes per polyp and 25 713 to 35 918 oocytes per colony. Male colonies of similar size to the female samples were shown to have 6 to 76 cysts per polyp and 14 014 to 27 019 cysts per colony. A. murrayi is a sessile gonochoric species with a 1:1 sex ratio and is most likely a broadcast spawner. The species has high fecundity, large eggs that could represent larger targets for sperm, primitive sper- matophores, as well as a large number of polyps per colony. These factors, along with a patchy distri- bution, would enhance the chance of fertilization for A. murrayi and may guarantee a successful reproductive strategy for this species.

KEY WORDS: Pennatulacea · Gametogenesis · Fecundity · Deep-sea

Resale or republication not permitted without written consent of the publisher

INTRODUCTION ever, data from a larger number of octocoral species is required for any robust analysis of reproductive Knowledge of the reproductive biology of organisms biology for this group. is essential to understanding how populations can be Information on reproductive traits of deep-sea corals maintained and renewed. Data comprising, for exam- is more recent and limited. Some deep-sea studies ple, the sexuality, reproductive mode, sex ratio and impose serious restrictions on specimen data acquisi- seasonality of reproduction of species of concern are tion and many studies use material already available in crucial to the establishment of better policies for the collections and are thus not ideally preserved. As such, management, conservation and recovery strategies of there are often few temporal series making it difficult degraded areas. to present conclusive investigations on timing of the Reproduction and propagation of shallow-water gametogenic cycle, duration of gametogenesis, sea- tropical corals have been extensively studied, particu- sonality and specific dates for gamete release or plan- larly for the scleractinians (Harrison & Wallace 1990). ulation. In spite of these logistical restraints, partially The reproductive biology of octocorals has also fo- accumulated data can show general trends of deep-sea cused primarily on tropical species (Simpson 2009), coral reproduction, and these are fundamental to max- and there is currently an initiative towards reviewing imizing outcomes and economizing effort when plan- the literature on this subject (Kahng et al. 2008). How- ning new sampling initiatives.

*Email: [email protected] © Inter-Research 2009 · www.int-res.com 104 Mar Ecol Prog Ser 397: 103–112, 2009

For general reproductive patterns in deep-sea corals, ing to 6 families, along the Brazilian coast. Seven of data exist primarily for groups such as scleractinians these recorded species came from depths below 200 m. (see review in Waller 2005), stylasterids (Brooke & The sea pen Anthoptilum murrayi Kölliker 1880 has Stone 2007) and octocorals (review in Simpson 2009). been reported from the shelf to the upper continental In terms of the sexual status and mode of development, slope of the North Atlantic, the northern Mid-Atlantic most shallow-water scleractinians are hermaphroditic Ridge, Faraday Seamount, the Indian Ocean and and spawn gametes for external fertilization (Harrison & waters around New Zealand and Australia (Jungersen Wallace 1990). In contrast, the limited available data on 1904, Thomson & Henderson 1906, Deichmann 1936, the deep-sea scleractinians to date shows that most are Cryer et al. 2002, McFadden et al. 2006, Molodtsova et gonochoric broadcast spawners. The exceptions are the al. 2008, Mortensen et al. 2008, Cairns et al. 2009). The species that have polyps adapted for brooding, such as first specimens from the South Atlantic were collected the Antarctic cup corals, Flabellum spp. (Waller et al. off Brazil, from 13° to 22° S (Pinto 2008). Most often, 2008) and 3 species of Caryophyllia which are herm- several specimens were sampled in a single dredge aphroditic (Waller 2005, Waller et al. 2005). (up to 52 colonies per dredge), which suggests high Twelve deep-sea stylasterids, belonging to the gen- local densities. era Stylaster, Errinopora, Distichopora, Cyclohelia and The present study describes the main characteristics Criptelia were studied by Brooke & Stone (2007). All on the reproductive biology of a member of the family were gonochoric brooders, in contrast to the shallow- Anthoptilidae, which has been considered one of the water Stylaster roseus, which may be hermaphroditic most primitive of the pennatulaceans (Doulan 2008). (Brooke & Stone 2007). The gametogenesis in male and female colonies of Despite the ecological importance of cold-water octo- Anthoptilum murrayi is described and estimates of corals, there are few published studies on their repro- their high fecundity are presented. Results of their ductive processes (Cordes et al. 2001, Orejas et al. 2002, reproductive characteristics are compared with data 2007), although some studies have included various as- available from other corals, particularly for sea pens. pects on the reproduction of pennatulaceans (Chia & As noted by Edwards & Moore (2009), A. murrayi con- Crawford 1973, Eckelbarger et al. 1998, Tremblay et al. firmed the pattern of both gonochorism and broadcast 2004, Soong 2005, Edwards & Moore 2008, 2009). How- spawning, which may be conserved at the order level ever, only 2 studies have focused on deep-water sea pens in Pennatulaceans. (Kophobelemnon stelliferum, Rice et al. 1992, and Um- bellula lindahli, Tyler et al. 1995). Gonochorism is the dominant pattern found for octo- MATERIALS AND METHODS corals, with only a few alcyonacean taxa being herm- aphroditic, e.g. Sinularia exilis (Benayahu 1997) and Colonies were collected off Bahia and Espírito Santo some species of Heteroxenia and Xenia (Benayahu States between 13° and 21° S in 2000 during the Living 1991, 1997). The frequency of brooding versus broad- Resources in the Exclusive Economic Zone (REVIZEE) cast spawning in octocorals varies with taxonomic Score Central Project (see Lavrado & Ignácio 2006), order (Simpson 2009). All pennatulaceans studied to and in 2003 off Rio de Janeiro State (Campos Basin, 21° date are broadcast spawners. to 22° S) during the Campos Basin Deep-Sea Environ- Anthozoans, especially scleractinians and octocorals, mental Project/PETROBRAS (Ocean Prof I and II cam- are the most dominant macrobenthic group in some paigns) (see Falcão et al. 2006). Specimens were up to deep-sea areas off Brazil (Castro et al. 2006). The pen- 75 cm in height (Fig. 1) and were collected at depths natulaceans are common octocorals, living in sediment between 1051 and 1799 m using otter trawls. A rela- areas surrounding deep-sea coral habitats. The order tively large number of specimens were collected dur- Pennatulacea occurs in all oceans, and species are gen- ing sampling, but only a few of these colonies (n = 24) erally adapted to living on soft sediments (Williams were undamaged and still bore intact polyps suitable 1995). Pennatulids are characterized by an oozooid (the for study. The 24 colonies were collected in February persistent and modified primary polyp), a proximal (4 specimens), June (9), July (4) and August (7). muscular peduncle and a distal polyp-bearing rachis Gametogenesis was examined using histological (polyparium) with autozooids (polyps with 8 well-de- methods. All colonies were fixed and preserved in veloped tentacles and mesenteries) and siphonozooids 70% alcohol, as they had been primarily collected for (polyps a with strongly developed siphonoglyph and taxonomic studies. The use of alcohol-preserved spec- reduced or absent tentacles) (Bayer et al. 1983). imens for histology was far from ideal and posed many The group has approximately 200 species (Williams, limitations during the various histological procedures. 1995). Castro & Medeiros (2001) and Castro et al. Tissues were fragile from alcohol preservation and the (2006) recorded 13 species of pennatulaceans, belong- process of extending the sections prior to application to Pires et al.: Reproduction of Anthoptilum murrayi 105

BH2 microscope. The longest axis of the oocytes and spermatic cysts were measured. Only perfect oocytes with nuclei were measured, to avoid remeasurement of the same cells. The polyparium of male and female colonies (from June 2000) was divided into 3 zones (distal, medial and basal) to evaluate differences in gamete development between zones. Polyps from each zone were examined using histology to access the synchrony of develop- mental stages among different colony zones. Polyps from different zones were also examined through dis- sections to determine polyp fecundity (= number of oocytes or spermatic cysts per polyp). The same colonies were used in both cases. Gametes of 4 polyps from each zone, as well as the total number of polyps in each colony, were counted in 4 male and 4 female colonies, all collected in June 2000 (total = 48 male and 48 female polyps). One of the male colonies was used solely for fecundity counts. Counts were made using a Zeiss SV6 stereomicroscope and all oocytes and spermatic cysts seen at 25 to 32× magnifi- cation were counted. Colony fecundity was estimated as the average number of gametes in polyps from dif- ferent zones of colonies multiplied by the average number of polyps of a given colony. Colony lengths were also measured. Sex ratio was tested using a chi-square comparing the observed and expected frequencies in a 1:1 ratio (null hypothesis). Such an expected ratio was desig- nated as the sum of males (M) and females (F) ob- served divided by 2 (n each sex = M/2 + F/2). Fig. 1. Anthoptilum murrayi. A deep-sea pennatulacean Differences in fecundity were tested using nested specimen collected off Brazil, southwestern Atlantic (Museu Nacional, Rio de Janeiro, MNRJ 4281). Po: polyparium; Pe: ANOVA, with polyp position nested in colony. Prior to peduncle; A: autozooid. Scale bar = 2 cm testing, homocedasticity of variance (Levene’s test) and normality (Kolmogorov-Smirnov test) were tested. No significant departures from ANOVA premises were the slides was difficult, with several sections lost dur- found, except for number of cysts (F11,36 = 2.59, p = ing reagent transfer. Polyps were dissected dehy- 0.016). Significant differences (p < 0.05) were further drated by graded alcohol series and embedded in examined with post hoc tests (Tukey’s HSD). All tests paraffin wax using standard methodology (Pantin were performed using Statistica 6.0 (StatSoft). 1948). It was ascertained that thinner sections were more adequate to obtain better histological results and as such sections of 4 to 6 µm were stained using Mal- RESULTS lory’s triple stain for histological observations. Gameto- genesis was classified according to histological charac- All examined colonies (females: 28 to 58 cm; males: 24 teristics and sizes of both the oocytes and spermatic to 64 cm long) presented gametes and were gonochoric. cysts. Stages of development are arbitrary as they It was not possible to distinguish the sex of the fixed reflect a continuous process (Wourms 1987). Gamete colonies without histological preparations. Sex ratio did development classification followed Pires et al. (1999), not significantly differ from 1:1 (11 males and 13 females where Stage I represented the beginning of develop- examined through histology, χ2 = 0.167, df = 1, p = 0.683). ment, Stage II an intermediate step (vitellogenic pro- Gametes appeared only in the autozooids. Mature cess in oogenesis) and Stage III mature oocytes and/or sexual cells were visible with the naked eye (Fig. 2). A spermatic cysts. Measurements of the different devel- layer of mesoglea surrounded the sexual cells and opmental stages of oocytes and spermatic cysts were stained an intense blue. The mesoglea thickened in the made using an eyepiece micrometer in an Olympus oocytes as they developed. 106 Mar Ecol Prog Ser 397: 103–112, 2009

Fig. 2. Anthoptilum murrayi. (A) Mature sperm cysts in a male colony. S: spermatic cysts. (B) Dissected polyp of a female colony showing oocytes in different stages of development. Scale bars = 1 mm

Oogenesis was classified into 3 stages (Fig. 3A–C). layers were inconspicuous in some cells, but usually Stage I represented primary development, Stage II an their thickness increased as the growth of sexual cells intermediate step (most of the vitellogenesis phase) advanced. Follicle layers were approximately 20 µm and Stage III characterized mature sexual cells. Early thick in sperm cysts and 70 µm thick around Stage III Stage I cells stained bluish purple. Pre-vitellogenic oocytes (Fig. 3C). Stage I cells usually presented a more homogenous Anthoptilum murrayi is more likely to present con- cytoplasm, and some cells stained light rose. The tinuous breeding activity than seasonal cycles, since nucleus was most often positioned centrally. The gametes in different stages of development were ob- smallest Stage I cell observed was 10 µm (longest axis) served from single seasons (Fig. 3B). The lack of con- and the largest was 350 µm, with a mean (± SD) value tinuous sampling meant that an estimate of the dura- of 115.32 ± 53.21 µm. Stage II oocytes stained dark tion of the reproductive cycle could not be obtained. rose. More lipid vesicles were present and had begun Polyps from different areas of the colony were shown to coalesce. Stage II oocytes showed a migration of the to bear gametes. In females, most of the oocytes were nucleus to the cell border and they ranged in size from in the earliest stages of development and polyps from approximately 220 to 540 µm, with a mean value of the basal area presented the highest frequency of very 368.28 ± 83.89 µm. Stage III oocytes stained from dark small cells (Fig. 4). rose to red. The cytoplasm became full of distinct color- High frequencies of small oocytes (up to 200 µm) and less lipid vesicles and the nucleus had moved to the low frequencies of large oocytes were observed for all periphery of the cell. Stage III oocytes ranged in size sampling periods (Fig. 5). All spermatic cysts observed from approximately 490 to 1200 µm, with a mean value in specimens collected in August were larger than of 901.07 ± 180.07 µm. 200 µm. Samples from August also showed the highest Spermatogenesis was also divided in 3 development frequencies of the largest size classes of cysts and the stages (Fig. 3D–F). Stage I spermatic cysts ranged from occurrence of the largest cysts observed for all samples 30 to 300 µm, with a mean value of 94.67 ± 54.17 µm. (between 700 and 740 µm) (Fig. 5). In Stage II, cysts had a lumen present and occasion- There were large numbers of gametes in both male ally some tails of spermatozoa could be seen. Stage II and female colonies. Examined specimens from June cysts ranged from 50 to 650 µm, with a mean value of 2000 had 414 to 645 polyps in female colonies (up to 299.03 ± 126.15 µm. In Stage III cysts, the heads of 747 mm in length) and 429 to 655 polyps in male colo- spermatozoa were located near the periphery of the nies (up to 650 mm in length). Female colonies pre- cyst and their tails projected into the center. Stage III sented 0 to 90 (47.6 ± 12.4) oocytes per polyp and an cysts ranged from 240 to 740 µm, with a mean value of estimated 25 713 to 35 918 (31 465 ± 5080) oocytes per 383.33 ± 85.67 µm. colony. Oocytes per polyp differed from polyp to polyp Male and female cells were associated with the folli- (ANOVA, F = 3.69, df = 8, p = 0.003). Tukey’s HSD cle layers (differentiated and flattened gastrodermal showed these differences always occurred between cells) during all stages of gametogenesis. These follicle the basal polyps and the middle or distal polyps (8 out Pires et al.: Reproduction of Anthoptilum murrayi 107

Fig. 3. Anthoptilum murrayi. Histology of the different stages of gametogenesis. (A) Stage I oocytes, (B) Stage II oocyte (2), but also showing a Stage I oocyte (1) and part of a Stage III oocyte (3), (C) Stage III oocyte, (D) Stage I sperm cyst, (E) Stage II sperm cyst and (F) Stage III sperm cyst. Fl: follicle cell layer. Scale bars = (A,C) 200 µm and (B, D–F) 100 µm of 32 such pairs). The number of oocytes per colony polyps presented significantly less cysts than middle also differed significantly from colony to colony and distal polyps (Tukey’s HSD, 19 out of 32 such com- (F = 7.33, df = 3, p < 0.0006). Male colonies of similar parisons). sizes presented 6 to 76 (36.6 ± 3.6) cysts per polyp and Anthoptilum murrayi is most likely a broadcast an estimated 14 014 to 27 019 (19 871 ± 5793) cysts per spawner, as no embryonic or planula stages were seen colony. The number of cysts differed significantly in the gastrovascular cavities of the polyps. In addition, among the different polyp positions on the colony large mature oocytes and intact cysts were commonly (F = 9.00, df = 8, p = 0.000001), but did not differ signif- seen above the pharynx, well up in the hollow tenta- icantly among colonies (F = 1.17, df = 3, p = 0.33). Basal cles. Fig. 6 shows a longitudinal section of a male polyp 108 Mar Ecol Prog Ser 397: 103–112, 2009

where mature cysts are seen in the tentacles and at the base of the polyp. It seems that intact cysts passed by the side of the pharynx and reached the tentacles.

DISCUSSION

Although the lack of a continuous sampling design imposed limitations on the present study, the available samples were sufficient to detect main features of reproductive biology for this species. Anthoptilum murrayi followed the same general pattern of gono- choric sexual development as seen in other pennatu- laceans and some octocorals (Table 1). The few avail- able data show that sea pens have a sex ratio of 1:1 (Edwards & Moore 2008, 2009), and this was observed in A. murrayi. The largest size of oocytes of Anthoptilum murrayi (1200 µm) was approximately 60% larger than the largest pennatulacean eggs recorded to date. The maximum oocyte size recorded for both Kophobelem- non stelliferum and Umbellula lindahli is 800 µm (Rice et al. 1992, Tyler et al. 1995), and Pennatula aculeata had a maximum oocyte size of 880 µm (Eckelbarger et al. 1998). Very large oocytes (up to 1200 µm) have also been recorded in other cold-water octocorals such as Dasystenella acanthina (Orejas et al. 2007), as well as in solitary deep-sea scleractinians belonging to the genus Flabellum (Waller et al. 2008). Increased oocyte size is correlated with some corals with an extended oogenic cycles, but not all corals with long cycles pro- duce large oocytes (Harrison & Wallace 1990). This size of oocyte also indicates that A. murrayi produces lecithotrophic larvae. In octocorals, larger oocytes are often associated with species which have these non- feeding larvae (Edwards & Moore 2009) and most planula observed appear to be lecithotrophic (Simpson 2009). Lecithotrophy is the pattern seen among penna- tulaceans (Eckelbarger et al. 1998) and is also ob- served in deep-sea broadcast spawning scleractinians (Waller 2005). Environmental factors influence coral sexual pro- cesses, synchronizing the reproductive cycles (affected by temperature, day length and salinity) and the tim- ing of spawning within a species, which is affected by tidal patterns and lunar rhythms (Harrison & Wallace 1990). However, the great majority of deep-sea species reproduce aperiodically or continuously, not requiring periodic environmental cues to regulate their gameto- genic cycles (Young 2003). Anthoptilum murrayi dif- Fig. 4. Anthoptilum murrayi. Relative frequency of oocyte size fers from most pennatulaceans, presenting a possible classes in polyps from the apical, medial and basal zones of continuous and long breeding activity rather than colonies. All colonies collected in June 2000. Key notation: colony zone (no. colonies/total no. gametes examined/mini- seasonal cycles. Continuous or quasi-continuous re- mum–maximum no. gametes examined in a single colony). productive cycles are not the general pattern in pen- Error bars are +SE natulaceans, but they have also been observed in Pires et al.: Reproduction of Anthoptilum murrayi 109

Fig. 5. Anthoptilum murrayi. Relative frequency of oocyte and spermatic cyst size classes in colonies collected in February, June, July and August. Key notation: month (no. colonies/total no. gametes examined/minimum–maximum no. gametes examined in a single colony). Error bars are +SE 110 Mar Ecol Prog Ser 397: 103–112, 2009

Fig. 6. Anthoptilum murrayi. Longitudinal section of a polyp showing intact Stage III spermatic cysts at the base of autozooids and at the tip of tentacles. S: mature spermatic cysts; T: tentacle; GC: gastric cavity. Scale bar = 600 µm

Table 1. Sexual pattern and spawning timing of Pennatulacea (data arranged in chronological order of source)

Species Locality Depth (m) Sexual pattern Spawning Source

Ptilosarcus Alki Point, Seattle, Shallow Gonochoric Late March Chia & Crawford (1973) guerneyi USA Kophobelemnon Porcupine Seabight, 350–1600 Gonochoric No signs of seasonal Rice et al. (1992) stelliferum SW Ireland reproduction Umbellula Porcupine Seabight, 650–3850 Gonochoric ? Tyler et al. (1995) lindahli SW Ireland Pennatula Gulf of Maine, USA 113–231 Gonochoric Suggests continuous Eckelbarger et al. (1998) aculeata spawning Renilla Southern California, Subtidal Gonochoric May–July Tremblay et al. (2004) koellikeri USA Virgularia Chiton Bay, Taiwan 0.5–1.0 Gonochoric August/ Soong (2005) juncea September? Pennatula W Scotland 18.2–19.9 Gonochoric July and/or August Edwards & Moore (2008) phosphorea Funiculina W Scotland 18.9–24.3 Gonochoric October–January Edwards & Moore (2009) quadrangularis Anthoptilum SW Atlantic, Brazil 1300–1799 Gonochoric Continuous Present study murrayi

Kophobelemnon stelliferum and Pennatula aculeata a different pattern was seen for 2 other deep-sea prim- (Rice et al. 1992, Eckelbarger et al. 1998; Table 1). noid species (Fanyella rossii and F. spinosa) in which The presence of oocytes in different stages of devel- reproductive cycles are annual. Cordes et al. (2001) opment in the same polyp can indicate extended and also observed continuous reproduction in the deep-sea continuous reproductive cycles, and also occurs in brooding alcyonacean Anthomastus ritteri, inferred by other deep-sea octocorals such as Antarctic primnoids the occurrence of gonads in all stages of development (Brito et al. 1997, Orejas et al. 2002, 2007). Brito et al. in the examined samples, as well as from the random (1997) identified a similar pattern for Thouarella vari- temporal pattern of planulation in the laboratory. abilis, and showed a 2 yr cycle of oogenesis or continu- While often not possible to obtain, additional monthly ous gametogenesis. Thouarella sp. and Dasystinella samples, or even those from smaller sampling period acanthina also have a long reproductive cycle with intervals, would be necessary to produce a more robust overlapping oocyte generations, each of which lasts estimate of oogenesis duration. more than 1 yr (Orejas et al. 2007). A gametogenic Polyps from different areas of the Anthoptilum mur- cycle of 18 mo, and possibly 2 yr, was also observed in rayi colony bear gametes. In females, most of the Ainigmaptilon antarcticum, and has been seen in other oocytes were in the earliest stages of development and deep-water anthozoans (Orejas et al. 2002). However, polyps from the basal area presented the highest fre- Pires et al.: Reproduction of Anthoptilum murrayi 111

quency of very small cells (Fig. 4). The large number of rence for Pennatula aculeata and suggested that early stage gametes has been reported in other octo- intact cysts could function as primitive spermato- corals, including the Pennatulacea (see Edwards & phores, potentially reducing sperm dilution. High Moore 2008). According to Brazeau & Lasker (1989), fecundity, large egg size (up to 1200 µm in histologi- the persistence of small oocytes throughout the year cal preparations) that could represent larger targets could indicate that oocytes require more than a single for sperm (Levitan 1996), primitive spermatophores season to mature. (sensu Eckelbarger et al. 1998) and large numbers of The duration of Stage III oocyte development of polyps per colony (up to 655 in examined A. murrayi) Anthoptilum murrayi is unknown as cells of this stage are all strategies that could enhance the chance of fer- have a wide size range. As such, the presence of large, tilization (Levitan 1996). There is also evidence of mature eggs does not necessarily indicate a spawning large, aggregated populations of A. murrayi (authors’ event is imminent. However, usually cyst development unpubl. data), as has been observed with other sea is faster and the spawning event is close when the tails pen species (Greathead et al. 2007). This would also of spermatozoa are present (Brazeau & Lasker 1989). increase the chance of fertilization in sessile gono- The final stages of coral spermiogenesis can proceed choric species with a 1:1 sex ratio. All these features very rapidly (Harrison & Wallace 1990). Most (58%) may guarantee a successful reproductive strategy in cysts examined in samples of A. murrayi collected in A. murrayi. August were Stage III, suggesting that the spawning event was close to this time, compared with 5% observed in June and none in February. There may be Acknowledgements. We thank the Assessment of the Sustain- other periods of sperm release during the year, but able Yield of the Living Resources in the Exclusive Economic data in the present study do not show this. A long Zone - REVIZEE project and the Campos Basin Deep-Sea Environmental Project/PETROBRAS for donating the speci- reproductive cycle, with a probable seasonal spawning mens for this study. Thanks to PETROBRAS and the United period, was also suggested to occur in Dasystinella Nations Environment Program (UNEP) for travel grants to acanthina and Thouarella sp. (Orejas et al. 2007). D.O.P. and C.B.C. to attend the 4th International Symposium Differences in fecundity among female colonies of on Deep-Sea Corals. D.O.P. and C.B.C. thank the Conselho Nacional de Desenvolvimento Científico e Tecnológico Anthoptilum murrayi should be interpreted with cau- (CNPq), Brazil, for research fellowships. We gratefully tion, as the differences among the polyps suggested it acknowledge MEPS editors and 3 anonymous referees for pro- would be necessary to sample more polyps for a more viding feedback that substantially improved the manuscript. robust analysis. Nevertheless, although fecundity was We also thank R. Waller for valuable comments on a revised roughly estimated, the results showed that the mean version of the manuscript and D. Tracey for some text revision. fecundity of female colonies was high (31 465 ± 5080 oocytes per colony). This figure may be an underesti- mate, since very small oocytes may have been over- LITERATURE CITED looked in the dissected samples. Total fecundity for A. Bayer FM, Grashoff M, Verseveldt J (1983) Illustrated trilin- murrayi is high and comparable with fecundity esti- gual glossary of morphological and anatomical terms mates in other pennatulacean species, which have applied to Octocorallia. Brill, Leiden ranged from approximately 30 000 to 200 000 oocytes Benayahu Y (1991) Reproduction and development pathways per colony (Edwards & Moore 2009). Frequency of of Red Sea Xeniidae (Octocorallia, Alcyonacea). Hydro- biologia 216-217:125–130 large oocytes was low in all examined samples (Fig. 5), Benayahu Y (1997) Development episodes in reef soft corals: suggesting that only a small percentage of oocytes ecological and cellular determinants. Proc 8th Int Coral mature at a time. Relatively high total fecundity, but Reef Symp 2:1213–1218 with a small proportion of developing oocytes attaining Brazeau DA, Lasker HR (1989) The reproductive cycle and spawning in a Caribbean gorgonian. Biol Bull 176:1–7 maximum size at a time, have been recorded in the sea Brito TAS, Tyler PA, Clarke A (1997) Reproductive biology of pens Umbellula lindahli and Funiculina quadrangu- the Antarctic octocoral Thouarella variabilis Wright & laris (Tyler et al. 1995, Edwards & Moore 2009). Studer, 1889. Proc 6th Int Conf Coelenterate Biology, The absence of embryos or planulae inside the gas- Nationaal Natuurhistorisch Museum, Leiden, p 63–69 tric cavities of all examined specimens suggested a Brooke S, Stone R (2007) Reproduction of deep-water hydro- corals (family Stylasteridae) from the Aleutian Islands, broadcast spawning mode of reproduction in Anthop- Alaska. Bull Mar Sci 81:519–532 tilum murrayi. To date there are no records of pennat- Cairns SD, Gershwin LA, Brook FJ, Pugh P and others (2009) ulacean species that are adapted for brooding. It Phylum Cnidaria – corals, medusae, hydroids, myxozoans. seems that male colonies of A. murrayi release intact In: Gordon DP (ed) New Zealand inventory of biodiversity. Volume 1. Kingdom Animalia: Radiata, Lophotrochozoa, cysts into the water, since they were seen above the Deuterostomia. Canterbury University Press, Christchurch pharynx level of the colonies close to the tentacle tips. Castro CB, Medeiros MS (2001) Brazilian Pennatulacea Eckelbarger et al. (1998) described a similar occur- (Cnidaria: Octocorallia). Bull Biol Soc Wash 10:140–159 112 Mar Ecol Prog Ser 397: 103–112, 2009

Castro CB, Pires DO, Medeiros MS, Loiola LL, Arantes RCM, Mortensen PB, Buhl-Mortensen L, Gebruk AV, Krylova EM Thiago CM, Berman E (2006) Cnidaria: Corais. In: Lavrado (2008) Occurrence of deep-water corals on the Mid- HP, Ignácio BL (eds) Biodiversidade bêntica da costa cen- Atlantic Ridge based on MAR-ECO data. Deep-Sea Res II tral brasileira. Museu Nacional, Rio de Janeiro, p 147–192 55:142–152 Chia FS, Crawford BJ (1973) Some observations on gameto- Orejas C, López-González PJ, Gili JM, Teixidó N, Gutt J, genesis, larval development and substratum selection of Arntz WE (2002) Distribution and reproductive ecology of the sea pen Ptilosarcus guerneyi. Mar Biol 23:73–82 the Antarctic octocoral Ainigmaptilon antarcticum in the Cordes EE, Nybakken JW, VanDykhuizen G (2001) Repro- Weddell Sea. Mar Ecol Prog Ser 231:101–114 duction and growth of Anthomastus ritteri (Octocorallia: Orejas C, Gili JM, López-González PJ, Hasemann C, Arntz Alcyonacea) from Monterey Bay, California, USA. Mar WE (2007) Reproduction patterns of four Antarctic octo- Biol 138:491–501 corals in the Weddell Sea: an inter-specific, shape, and Cryer M, Hartill B, O’Shea S (2002) Modification of marine latitudinal comparison. Mar Biol 150:551–563 benthos by trawling: Toward a generalization for the deep Pantin CFA (1948) Notes on microscopical techniques for ocean? Ecol Appl 12:1824–1839 zoologists. Cambridge University Press, Cambridge Deichmann E (1936) The Alcyonaria of the western part of the Pinto RP (2008) Novo registro da família Anthoptilidae Atlantic Ocean. Mem Mus Comp Zool Harv 53:1–317 (Cnidaria: Octocorallia: Pennatulacea) no Atlântico Sul: Doulan E (2008) Phylogenetics, systematics and biogeogra- Anthoptilum murrayi Kölliker 1880. BS dissertation, Uni- phy of deep-sea Pennatulacea (Anthozoa: Octocorallia). versidade Santa Úrsula, Rio de Janeiro Evidence from molecules and morphology. PhD thesis, Pires DO, Castro CB, Ratto CC (1999) Reef coral reproduction University of Southampton in the Abrolhos Reef Complex, Brazil: the endemic genus Eckelbarger KJ, Tyler PA, Langton RW (1998) Gonadal mor- Mussismillia. Mar Biol 135:463–471 phology and gametogenesis in the sea pen Pennatula Rice AL, Tyler PA, Paterson GJL (1992) The pennatulid aculeata (Anthozoa: Pennatulacea) from the Gulf of Kophobelemnon stelliferum (Cnidaria: Octocorallia) in the Maine. Mar Biol 132:677–690 Porcupine Seabight (north-east Atlantic Ocean). J Mar Edwards DCB, Moore CG (2008) Reproduction in the sea pen Biol Assoc UK 72:417–434 Pennatula phosphorea (Anthozoa: Pennatulacea) from the Simpson A (2009) Reproduction in octocorals (Subclass Octo- west coast of Scotland. Mar Biol 155:303–314 corallia): a review of published literature. Available at Edwards DCB, Moore CG (2009) Reproduction in the sea pen www.dmc.maine.edu/sites/watlingsite/PAGES/repro.html Funiculina quadrangularis (Anthozoa: Pennatulacea). Soong K (2005) Reproduction and colony integration of the Estuar Coast Shelf Sci 82:161–168 sea pen Virgularia juncea. Mar Biol 146:1103–1109 Falcão APC, Mauro CA, Cavalcanti GH, Paes JES and others Thomson JA, Henderson WD (1906) An account of the Alcy- (2006) Biodiversidade marinha brasileira; algumas con- onarians collected by the Royal Indian Marine Survey Ship tribuições da Petrobras. Petrobras, Rio de Janeiro Investigator in the Indian Ocean. Part I. The Alcyonarians of Greathead CF, Donnan DW, Mair JM, Sounders GR (2007) the deep-sea. The Indian Museum, Calcutta, p 1–132 The sea pens Virgularia mirabilis, Pennatula phosphorea Tremblay ME, Henry J, Anctil M (2004) Spawning and and Funiculina quadrangularis: distribution and conser- gamete follicle rupture in the cnidarian Renilla koellikeri: vation issues in Scottish waters. J Mar Biol Assoc UK 87: effects of putative neurohormones. Gen Comp Endocrinol 1095–1103 137:9–18 Harrison PL, Wallace CC (1990) Reproduction, dispersal and Tyler PA, Bronsdon SK, Young CM, Rice AL (1995) Ecology recruitment of scleractinian corals. In: Dubinsky Z (ed) Eco- and gametogenic biology of the genus Umbellula (Pennat- systems of the World 25. Coral reefs. Elsevier, Amsterdam, ulacea) in the North Atlantic Ocean. Int Rev Gesamten p 133–207 Hydrobiol 80:187–199 Jungersen HFE (1904) Pennatulida. The Danish Ingolf Ex- Waller RG (2005) Deep-water Scleractinia (Cnidaria: Antho- pedition, Vol 5. Bianco Lunus Bogtrykkeri, Copenhagen, zoa): current knowledge of reproductive processes. In: p1–95 Freiwald A, Roberts JM (eds) Cold-water corals and eco- Kahng S, Benayahu Y, Lasker H (2008) A review of sexual systems. Springer-Verlag, Berlin, p 691–700 reproduction in Octocorallia. Proc 11th Int Coral Reef Waller RG, Tyler PA, Gage JD (2005) Sexual reproduction in Symp, 7–11 Jul 2008, Fort Lauderdale, FL, p 98 (Abstract) three hermaphroditic deep-sea Caryophyllia species Kölliker AV (1880) Report on the Pennatulida dredged by (Anthozoa: Scleractinia) from the NE Atlantic Ocean. H.M.S. Challenger during the years 1873–76. In: CW Coral Reefs 24:594–602 Thomson, J Murray (eds) Report of the scientific results of Waller RG, Tyler PA, Smith CR (2008) Fecundity and embryo the voyage of H.M.S. Challenger during the years 1873–76. development of three Antarctic deep-water scleractinians: Zoology Vol 1, Part 2, p 1–41 Flabellum thouarsii, F. curvatum and F. impensum. Deep- Lavrado HP, Ignácio BL (2006) Biodiversidade bêntica da Sea Res II 55:2527–2534 costa central brasileira. Museu Nacional, Rio de Janeiro Williams GC (1995) Living genera of sea pens (Coelenterata: Levitan DR (1996) Effects of gamete traits on fertilization in Octocorallia: Pennatulacea): illustrated key and synopses. the sea and the evolution of sexual dimorphism. Nature Zool J Linn Soc 113:93–140 382:153–155 Wourms JP (1987) Oogenesis. In: Giese AC, Pearse JS (eds) McFadden CS, France SC, Sánchez JA, Alderslade P (2006) A Reproduction of marine invertebrates, Vol IX. General molecular phylogenetic analysis of the Octocorallia aspects: seeking unity in diversity. The Boxwood Press, (Cnidaria: Anthozoa) based on mitochondrial protein-cod- Palo Alto, CA, p 49–178 ing sequences. Mol Phylogenet Evol 41:513–527 Young CM (2003) Reproduction, development and life-history Molodtsova TN, Sanamyanb NP, Keller NB (2008) Anthozoa traits. In: Tyler P (ed) Ecosystems of the World 28. Eco- from the northern Mid-Atlantic Ridge and Charlie-Gibbs systems of the deep oceans. Elsevier, Amsterdam, Fracture Zone. Mar Biol Res 4:112–130 p 381–426

Submitted: March 2, 2009; Accepted: September 16, 2009 Proofs received from author(s): October 25, 2009